We assume the ledger is honest, but the numbers moving across it answer to someone else's interest rate. That assumption is the thing I keep circling back to, because every stablecoin dashboard I have ever opened presents supply as a story of adoption—a line going up and to the right, driven by payments, remittances, and DeFi collateral demand. None of them present supply as a function of the federal funds rate. A recent BIS working paper suggests the second framing may be the more honest one: when US banks raise deposit rates, stablecoin growth slows. The finding is presented as a modest academic observation. Read structurally, it is something closer to a confession that the fastest-growing dollar instrument of the last decade has a thermostat, and the thermostat is not in the crypto industry's control.
Liquidity is a mirage. I have written that line for years about DeFi depth, but it applies with more force here. The paper's central move is to model stablecoin demand as a substitute for retail bank deposits rather than as a speculative crypto asset. That is not a neutral technical choice—it is a reclassification with regulatory consequences baked in. Once stablecoins are deposits in all but name, the appropriate supervisory path shifts from securities or commodity frameworks toward banking and payment oversight. The intellectual framing precedes the policy outcome, always.
The Composition Effect Nobody Bothers to Check
The headline finding that USDC is the most rate-sensitive stablecoin deserves scrutiny rather than repetition. My instinct after auditing enough on-chain flows is that sensitivity differentials across issuers are almost never pure elasticity parameters. They are composition effects wearing a lab coat.
Consider who actually holds each instrument. USDC's user base skews toward US-domiciled institutions, funds, and DeFi protocols whose operators maintain bank accounts and treasury functions. These holders possess a genuine menu: park dollars in a money market fund at five percent, or hold zero-yield USDC for on-chain liquidity. When deposit rates rise, that menu becomes expensive to ignore. Now consider USDT's holder base, which is heavily concentrated in emerging markets and capital-controlled jurisdictions where local deposit rates are not a meaningful alternative to a dollar-denominated instrument—they are a different currency entirely, and often a depreciating one. A Nigerian importer holding USDT for settlement is not choosing between Tether and a Treasury bill; the comparison set does not exist.
This means the paper's "USDC is most sensitive" conclusion may be measuring holder demographics rather than rate elasticity as a causal parameter. The distinction matters enormously for anyone building a forecast. If it is demographics, then USDC's sensitivity will decline as its user base globalizes, and Tether's apparent immunity is a function of its geography, not its architecture. If it is true elasticity, then USDC's growth ceiling is structural and permanent. The source material I am working from does not disclose the identification strategy, the sample window, or the coefficient values—only the directional claim. That gap is not a minor academic detail. Without the causal identification, the entire finding collapses into the observation that crypto activity correlates with risk appetite, which we already knew.
The Reverse Exposure That Breaks Valuation Models
Here is the layer the coverage missed entirely, and it is where I think practitioners should focus. Rate sensitivity cuts in two directions simultaneously, and the two directions move opposite to each other.
When rates rise, stablecoin supply growth decelerates—that is the paper's claim. But the same rate increase lifts the yield on the issuer's reserve portfolio, which is composed overwhelmingly of cash and short-duration Treasuries. When rates fall, supply growth accelerates while unit reserve income compresses. Volume and price react to the same variable in opposite directions, which means the net effect on issuer revenue is genuinely ambiguous and depends on reserve duration management rather than on growth. For an issuer like Circle, whose revenue is dominated by reserve interest income, this is not a footnote. It is the entire business model.
The practical implication is uncomfortable for anyone valuing these businesses on supply growth alone. A rate-cutting cycle will produce beautiful supply charts and potentially deteriorating earnings quality per dollar of float. A hiking cycle will do the reverse. Market consensus treats stablecoin supply growth as monotonically structural—a penetration story. If the BIS framing is even partially right, a meaningful component of that growth is cyclical and rate-driven, which means extrapolating a three-year adoption curve into a decade-long forecast embeds a monetary policy assumption nobody has stated.
I have watched this exact error before. In 2020 I spent months tracking Aave v2's isolated risk modules and the uncollateralized lending that appeared alongside them, and the pattern was identical: participants modeled an incentive structure as a permanent feature of the landscape when it was in fact a function of a rate environment that could reverse in a quarter. The architecture survived; the assumptions did not.
The Institutional Asymmetry
The paper lands in a policy environment where the conclusion has already been partially legislated. Under the current US stablecoin framework, issuers are prohibited from paying interest to holders. Read that provision again with the BIS finding in hand. It is not a technical consumer-protection measure. It is a permanent handicap written into law, preventing stablecoin issuers from defending their demand base when deposit rates rise. They cannot share reserve yield with holders to blunt the opportunity cost of holding a zero-yield instrument.
That is what "banks hold the leverage" actually means in operational terms. The mechanism is not market competition—it is the absence of a competitive tool. And it points toward where the real structural threat sits. Tokenized deposits—bank liabilities represented on distributed ledgers, still carrying deposit insurance, still inside the regulatory perimeter, and crucially able to pay interest—would occupy the same programmable settlement niche while possessing the one capability stablecoin issuers are legally denied. If the BIS logic is adopted into policy debate, the likely outcomes are scale caps, reserve-retention requirements, or a loosening of the interest prohibition for large institutional instruments. All three reshape the competitive field, and none of them appear in the standard stablecoin bull case.
What gets systematically misread, and I want to be precise about this, is the status of the source. A BIS working paper is a preprint. It carries no legal force, represents no official institutional position, and has not been peer reviewed. Its transmission path runs from research circles through central banks to finance ministries and legislatures over a horizon of six to eighteen months, and its real function in that chain is often to supply legitimacy for positions already held. Code is law, but who writes the law? In monetary infrastructure, the question is answered by whoever controls the research pipeline feeding the legislators.
The bear market context sharpens rather than softens this reading. When capital is scarce, the analytical question shifts from "how fast does this grow" to "what constrains it." Stablecoin supply is not a parallel universe with its own physics. It is a dollar liability sitting inside a rate regime it does not set, serviced by custodial banks whose failure transmits directly into peg risk—as holders learned in March 2023 when USDC briefly traded near eighty-seven cents. The demand-side rate sensitivity is gentle and gradual. The reserve-side and institutional risks are abrupt. Conflating them produces mispriced risk in both directions.
Your data is not yours anymore—and increasingly, your stablecoin float is not yours either. It is a rate-sensitive claim on a banking system that has both the tool and, evidently, the legislative protection to reclaim it. Watch the next easing cycle not for the supply chart, which will look magnificent, but for the divergence between float growth and per-unit reserve income. That gap is where the honest version of this story will be written.
